- Overview
- Broadcast Model
- Interactivity
- Data Delivery
- Authoring Content
- Packaging Content
- References
This chapter is from the book
This chapter is from the book
This chapter is from the book
3.5 Authoring Content
As opposed to traditional TV, where the content included video and audio carried in a single format, iTV content is composed of numerous types of content encoded in various formats. Today's Internet uses numerous file formats, including those listed in Table 3.3.
Table 3.3. Video File Extensions Found on the Internet
|
Application |
File Extension |
|---|---|
|
AVI files |
.avi |
|
RealAudio files |
.ra, .rm |
|
Secure RealMedia files |
.rms |
|
MP3 files |
.mp3, .pls, .m3u |
|
MPEG files |
.mpg |
|
RealJukebox files |
.rmx, .rmj |
|
AAC files |
.acp |
|
Music files |
.mes |
|
Liquid Audio files |
.lqt |
|
MJuice files |
.mjf |
|
IBM EMMS files |
.emm |
|
Windows Media Files |
.wma |
|
Blue Matter Files |
.bmo, .bmr, .bmt |
Video editing is a complicated, iterative process in which previously captured video segments, pulled from a library, are assembled together with other segments with the help of visual effects and graphic annotations. A storyboard is used to organize the various video segments and annotations along a timeline. Transitions from one video segment into another may be achieved through transition effects. Graphic animations, such as animated titles or figures, can be added at specified time points to provide information or partition the resulting composed video into sections. Background music may be mixed on top of the audio track recorded with the video. Various intermediate versions may be saved to files until a satisfactory version emerges.
The notion of the current movie is similar to a project concept and refers to the assembly of components currently edited. A library option is typically used to access video components that are previously recorded video content. Components can be loaded by dragging and dropping thumbnail images representing these components onto a timeline. Once a library component is loaded, it can be added as a whole or in parts.
A video timeline is the reference against which all components of the current movie are placed and synchronized. Each component of the movie is associated with a bar whose endpoints indicate the time points at which the component appearance begins and ends. Whereas some systems allow users to specify the starting and ending point of a clip's inclusion, other systems require the generation of an intermediary file containing the video clip cut at the desired locations.
Video clips may be added with overlapping time periods. At any point in time at which two or more video clips overlap, there is a need to specify how they will be mixed or otherwise combined. Most video editing software provides a library of effects that can be used to define transitions (see Figure 3.19). As an example, the new video clip may be introduced gradually or abruptly, using alpha blending, fade-in, fade-out, dissolved, wipe, push, cover, or reveal transitions.
)
Figure 3.19. A transition control user interface courtesy of Applied-Magic (applied-magic.com).
In addition to transitions, highly imaginative video sequences can be created using pan, zoom, and rotate tools. Specific examples of tools that can be used include zoom in, zoom out, pan up, pan down, zip, zigzag, and spinning of photographic images.
Titles and images over video are commonly used for iTV programming. Typically, a special user interface is provided for each type of annotation. Text titles, for example would be added and controlled with a screen that includes font, color, style, and size controls (see Figure 3.19). Images, for example, would be added with a screen that includes scaling, rotation controls, and possibly image editing controls.
The appearance of titles and images is controlled by transition effects of a different kind. Typical transition effects includes options for text to scroll up from the bottom or top, left or right, to dissolve, rotate, or to zoom in or out.
In many situations, adding background music or other audio tracks improves the viewer's experience. Much like video editing, authoring of Audio content is performed using various tools, capable of cutting, pasting, sequencing, and processing raw recorded audio as well as mixing it with Musical Instrument Digital Interface (MIDI) content. Basic editing tools feature the capability to handle data of arbitrary sample rate, 8-bit and 16-bit encoding, both mono and stereo. Advanced tools typically include numerous mixing and pattern options for combining multiple audio tracks, as well as libraries of hundreds of audio effects. An important capability is loading and processing multiple file formats, such as AIFF, AU, WAV, and sometimes, MP3.
MIDI was introduced in 1983. It was developed in cooperation between the major music industry electronic instrument manufacturers including Roland, Yamaha, Korg, and others. The protocol allows electronic devices (usually synthesizers, but also computers, light show controllers, VCRs, multitrack recorders, etc.) to interact and work in synchronization with other MIDI-compatible devices. Using a master controller device such as a keyboard, the author of an audio track can play or trigger sounds from other electronic devices remotely. This enables MIDI to offer faster creation and composition of music, rendering the composer a proverbial one-man band. As a result, throughout its existence, MIDI has gained acceptance among industry professionals.
3.5.1 Markup Content Type
Markup content type has emerged as a new type of content, complementing procedural content. It is a text-based format that specifies a document's hierarchical tree structure. As an example, a document may be organized into sections, each of which is organized into paragraphs, each of which may contain references, figures, and other special content. The beginning of a section could be marked with the <section> string, called a tag, in which case the end of that section must be marked with the </section> string (notice the '/').
One of the most common and visible applications of markup content is for purposes of specifying layout. The HTML was developed to enable specifying the layout of a document; that layout rendered automatically by an HTML interpreter commonly known as a Web browser. Other common visible applications of markup content type include archiving and search engines and digital signatures.
Today, there are numerous HTML-authoring tools. Most tools use templates, style sheets, customizable toolbars, and JavaScript wizards. Most word processing programs can produce HTML output, and some can edit HTML directly. Some advanced tools enable adding animation. For non-HTML applications, there are a number of smart text editors which recognize tags and introduce tag-specific functionality. All in all, there are countless Internet resources for authoring HTML and other markup content.
3.5.2 Script Content Type
The European Commerce Applications (ECMA) Script was designed to extend declarative markup content with a procedural language capability. This extends the expressive power of declarative markup languages to the point it is as powerful as any other programming language available. ECMA Script is based on several originating technologies, the most well known being JavaScript (Netscape) and JScript (Microsoft).
One of the most common applications of ECMA Script is event-based Dynamic HTML (DHTML) behaviors. It enables converting a static HTML page into an animated and interactive TV application. The DHTML concept is about grabbing user interface event to alter the content of a page after it has been loaded into the browser, in broadcast mode, namely without calling for action from the server. DHTML provides full local interactivity; there is no need for any server-side scripts nor is there a need to access any server.
Script content authoring is largely associated with DHTML authoring. There are a few automated script generators that are part of HTML template generators. Most authoring is performed using text editors, manually linking HTML using event capture attributes such as onclick() and onchange() (see Chapters 5 and 6 for details). All in all, there are countless Internet resources for authoring DHTML.
3.5.3 Java Content
Common to all iTV standards is the inclusion of Java. Java is a general-purpose object-oriented programming language. It improves on C and C++ by avoiding many of the features that make them confusing and unsafe. Java was designed to render code mobile by enabling objects to behave predictably when executed within unknown environments. In the context of iTV, objects authored production environments are compiled into class files and transported into an iTV receiver serving as the execution environment for those class files.
iTV Java content is essentially a collection of class files, one of which is an entry point implementing the JavaTV Xlet interface. Executable Java class files are compiled from a text-based source code into Java class files that are executed at the receiver by a Java virtual machine. JavaTV is a set of API that extends the general purpose Java platform with iTV specific functionality. Although Sun provides a reference implementation, receiver manufacturers are free to develop their own implementations of the JavaTV API specification.
Authoring Java applications is usually performed using Integrated Development Environments (IDE). These include a sophisticated text editor that integrates with a Java compiler, debugger, and JavaDoc documentation generator. There are a number of iTV-specialized IDEs supporting multimedia-integration features such as timeline synchronization.
3.5.4 Captions
Most broadcast media define a way for data service text to be delivered with the video signal. In some systems, this is called closed captioning or text mode service; in other systems, it is called teletext or subtitling.
Just as a caption is the text under a picture, captions on video are text located somewhere on the picture. Because there is no way for a television to put text outside the area of the picture tube, captions do end up covering a portion of the picture (there are nonbroadcast applications in which this can be done).
In contrast to captions, intended for the hearing impaired, subtitles are intended for hearing audiences. Subtitles summarize the audio and rarely report all the events. For example, captions report on sound effects such as phone ringing and footsteps, whereas subtitles do not.
Closed captions are captions that are hidden in the video signal, invisible without a special decoder. They are hidden in line 21 of the Vertical Blanking Interval (VBI). Open captions are captions that have been decoded, so they have become an integral part of the television picture, like subtitles in a movie. In other words, open captions cannot be turned off. The term open captions is also used to refer to subtitles created with a character generator.
Captions are available in terrestrial, satellite, and cable broadcasts, as well as DVD formats. Decoders were originally designed to allow for captioning in more than one language, although this is not done much (60 Minutes, which is captioned in English and Spanish, is an exception). Line 21, where the captions are carried, is split into two fields. The first field carries two caption channels, CC1 and CC2. The second field carries the other two, CC3 and CC4. Most external decoders and TV sets with built-in Line 21 decoders have an option for text and an option for captions. The text option, instead of displaying a few lines of captions somewhere on the picture, takes over all or half of the screen to display scrolling text information.
3.5.4.1 Online (Live) Captions
Online captioning is done in real time, as events occur or from a script. Television news shows, live seminars, and sports events are prime examples of content for which online captions are common. Such captions are prepared by stenocaptioners, who are trained to use the same shorthand keyboard a court reporter uses. In fact, most stenocaptioners began their careers as court reporters. They listen to the live broadcast at the same time viewers do and write what they hear at 250 words per minute or faster. Special computer software translates their phonetic shorthand into English.
Live display captions are typically sent to the TV station using a modem or an Internet connection. It is received by a closed-caption encoder, which places it on the broadcast signal in real-time, within a couple of seconds.
3.5.4.2 Offline (Post-Production) Captions
Offline captioning is done after the fact, in a studio. Examples of offline captioning include television game shows, videotapes of movies, and corporate videotapes (e.g., training videos). The text of the captions is created on a computer, and synchronized to the video using time-codes. The captions are then transferred to the videotape before it is broadcast or distributed. Whereas online captions require writing at breakneck speeds and result in errors, offline captioning, or stenocaptioning, allows correcting of the existing captions, or creating new captions, for an entire broadcast.
3.5.4.3 Hybrid Solutions
Many newscasts use a combination of captioning techniques to try to achieve both the accuracy of live display captioning and the complete coverage of real-time stenocaptioning. To accomplish this, the stenocaptioner dials in to the newsroom computer system about an hour before the broadcast and copies all of the scripts into the captioning system. The captioner then sorts and cleans up the scripts, names the segments, marks which ones will require live stenocaptioning, and prepares to go on the air.
During the broadcast itself, the stenocaptioner may jump back and forth between feeding out scripts and writing in real-time dozens of times. Real-time captions appear, with no delay, one word at a time, and scripts appear one line at a time but with some delay.
3.5.4.4 Applications
Whereas captions may have recreational value (to people in noisy bars or those who are on the phone), a number of studies have shown that captioning is an invaluable aid to teaching people to read, especially in the context of English as a second language. The national captioning institute reports that about half of all caption decoders have been sold to foreign immigrants learning English as a second language. People learning English for the first time find it easier to understand (and learn) idioms and other expressions when they see them in print as well as hear them. “Adult Literacy: Captioned Videotapes and Word Recognition” by Dr. Benjamin Michael Rogner explored the use of captions as a method of improving sight word recognition and teaching english as a second language. “Closed Captions: An Untapped Resource in Combating Illiteracy” by Andrea Shettle recommends turning on the caption decoder whenever young children and adults who have been struggling to learn how to read are in the room.
3.5.5 iTV Productions
TV content is grouped on an episode basis, and typically changes from one episode to the next. An episode contains items such as trivia questions and answers, location id's, points, duration, images, hyperlinks, etc. Within episodes, iTV applications include facilities for aggregation of users responses (e.g., leaderboards, polls), response to viewer selection action (e.g., trivia, incremental points), timing information for the synchronized display of media, and a combination of these facilities.
3.5.5.1 Production Process
The stages of the production process include (a) script writing, (b) video editing, and (c) production. There is live production as well as production after the master tape is produced. The production process produces content assets, each having the components of a content type, timing, and content body. There are a variety of different content types, related to games, polls, interactive advertisements, leaderboards, and more. The timing component specifies the time in the program at which a content item appears and the duration for which that item is displayed. The content body refers to the actual text and images that make up a content item. Each component could be determined at different stages of the production process. For example, the following scenarios are feasible:
-
Content type and body are determined in the script writing process while timing is finalized in the tape creation process.
-
Content type and body are determined in the script writing process while timing is finalized in the pre-airing process (after the tape has been finalized).
-
Timing and type are determined in post-production through the insertion of interactive markers in the video. The body is created later, based on the shell timing and type.
-
Content type, body, and timing are all created at the same time in the tape creation process.
-
Content body and type are pre-scripted and timing is determined in a live environment.
-
Content body, type, and timing are all produced in a live environment.
-
The interactive show is completely produced after the TV show has been finalized.
3.5.5.2 Presentation and Behavior
While content changes from episode to episode, the characteristics of the program remain the same for an entire season of episodes. These characteristics, referred to as presentation description, cover everything related to the application specific logic and the general look and feel of an interactive program, including elements such as location options for interactive assets, type of interface (e.g., right-side L-shape, left-side L-shape, overlay in bottom), fonts, sizes, etc.
3.5.5.3 Integrating Data and Video
The integration of data and video requires a sophisticated powerful authoring tool that can add data on top of prerecorded audio and video. The data to be integrated includes static text and images, markup content, script content, and Java applications. The authoring model should be based on story boards and timelines. The authoring tool should provide a simulator of the target receiver including the remote control (see Figure 3.20).
)
Figure 3.20. AltiComposer's MHP environment simulator courtesy of AltiCast (www.alticast.com).
As a basic functionality, the iTV authoring tool must enable management of scenes and scene components (see Figures 3.21 and 3.22). Each scene is associated with a number of parameters, such as starting times and transitions. Another basic capability is placing text and images over the video, at specific locations and at specific time points. Typically, placement is performed through drag-and-drop operation. With more sophisticated tools, the placement properties include blending options as well as transition and animation options. Other important features include the capability to mix-and-match off-the-shelf components available in large libraries, automatic generation of iTV DHTML content, and availability of actors and effects libraries.
)
Figure 3.21. Alticast composer's editor enables controlling where and when each object appears.
)
Figure 3.22. Alticast Scene Editor enables editing properties of each scene component.
According to Alticast's experience, the authoring process can be divided into six steps: initial preparation, scene structuring, adding interactivity, adding transitions, adding effects, and finally, testing.
3.5.5.4 Initial Preparations
In this step, the author prepares the project files, preliminary layout, and structure. This step also includes the identification of the libraries to be used, including prerecorded audio and video, template, image, effect, and transition libraries.
Preparation might also require the construction of custom components. This includes modification or composition of existing images, effects (sounds, animation, transitions), and actors to produce new application specific versions of these components. The resulting components, together with the prerecorded video and audio, are typically assembled to form the iTV program's resource library.
3.5.5.5 Scene Structuring
This step involves identifying the scenes and the relationships between them. For each scene identified and created, there is a need to create its list of planes, and for each plane, its list of shots (see Figure 3.23). The scene structure is then populated with components from the custom resource library prepared in Step 1. The scene editor should provide good control over each component and should enable previewing the scene as it would be shown in the final version (see bottom of Figure 3.20).
)
Figure 3.23. An example scene structure.
3.5.5.6 Interactivity, Transitions, and Effects
Next, interactivity is added to allow navigation between the various shots (i.e., screens) of the application. This includes creating interactive menus and selectable icons. The authoring tool should provide a remote-control simulator that enables previewing the functionality of the resulting application exactly as it would operate when deployed.
A critical aspect of this step is the authoring of transitions, which is the key factor impacting viewer experience and brand an application's look and feel. In some cases, when selections imply the launching of a video stream, it is critical to synchronize these transitions with the launched video.
After adding transitions, creative effects could be added. This step involves integration of sound effects and graphic animation and provides one of the best opportunities to apply creativity. For example, when the application transitions from one shot to the next, there is an opportunity to add an animation object accompanied by sound effects.
3.5.5.7 Preview and Testing
Preview and testing occur iteratively throughout the creative authoring process. This involves both examination of shots, and replay of the scene or the entire application. An authoring environment should provide an emulator of the target platform that includes correct mapping of all of the remote control keys and functions; in some cases, to improve the quality of the preview, a physical remote control and high-fidelity stereo system may be provided as well. The author should use the emulated (or physical) remote control to interact with the application and evaluate its look and feel. It is during this high-quality preview step that many creative ideas arise.